The B2B SaaS lead generation funnel

Funnel diagrams are everywhere and almost none of them help. What a marketing lead actually needs is a way to open last quarter’s numbers and identify, in about an hour, which stage is losing the pipeline. That requires exact stage definitions first, because a ratio between two fuzzy stages is a number without a meaning.

The six stages and what each event has to be

Six stages, each defined by a thing that happened at a timestamp, not by a feeling about intent.

Visitor. A session on your site. Simple, and the only stage most teams measure correctly.

Hand raiser. Submitted any form. Newsletter, ebook, demo request, pricing enquiry. This stage exists specifically so you stop pretending a whitepaper download and a demo request are the same thing.

MQL. Crossed a defined threshold combining fit and behaviour. Fit is firmographic: company size, industry, region. Behaviour is what they did. The threshold must be a number someone can point at.

SAL. A rep looked at it and agreed it’s worth contacting. This stage is where marketing’s claims meet sales’ opinion, and the acceptance rate is the most diagnostic single number in the whole funnel.

SQL. A rep spoke to them and confirmed real need, rough timing and some path to budget. Conversation happened.

Opportunity. A deal record with an amount and a close date in the CRM. Forecast exists.

Why published conversion benchmarks are useless without their event

Site conversion rates in circulation run from 1.1 to 7.6 percent, a seven fold spread that has nothing to do with the sites and everything to do with counting.

A benchmark counting any form submission, including newsletter and content downloads, lands near the top. One counting only demo requests lands near the bottom. Then the denominator moves too: sitewide traffic including support docs, blog readers and job applicants gives a very different number to landing page traffic only.

Stage transition	Common range	What moves it most	Diagnostic value
Visitor to hand raiser	1.1 to 7.6%	Traffic intent and offer match	Low, depends on event definition
Hand raiser to MQL	20 to 45%	Scoring threshold and fit filter	High
MQL to SAL	30 to 70%	Definition agreement with sales	Very high
SAL to SQL	25 to 45%	Speed to first contact and targeting	High
SQL to opportunity	45 to 70%	Qualification rigour	Medium
MQL to opportunity, blended	8 to 20%	Lead source mix	High, if segmented by source

Before you compare yourself to any of those, segment by source. Review site leads and branded search leads convert to opportunity at two to four times the rate of gated content leads. A blended MQL to opportunity number tells you your channel mix and very little else. Our SaaS lead conversion benchmarks and B2B SaaS funnel conversion rate benchmarks pages keep the cuts separate.

Finding the leak, in order

The method is mechanical and takes about an hour with clean data.

Diagnostic one: traffic rich, lead poor

Symptom: 40,000 monthly sessions, 120 hand raisers, 0.3 percent conversion. Everyone proposes a CRO project.

It’s almost never CRO. Pull your top twenty landing pages by traffic and check what someone searching those terms actually wants. If your traffic comes from definitional and top of funnel content, that audience isn’t buying anything today and no button colour changes that. You have a demand capture gap sitting behind a demand generation success.

The fix is more bottom of funnel surface area: comparison pages, alternatives pages, integration pages, pricing content. Then match the offer to the page. A lead magnet on a “what is X” article should be a template or a tool, not a demo request, and the demo request belongs on the comparison page where the buyer is already choosing.

The failure mode here is real and worth stating. Rebuilding your top of funnel content into bottom of funnel content shrinks your traffic number, sometimes by a lot, while improving your lead number. Somebody will present the traffic decline as a problem at the next board meeting. Get ahead of that before you start.

Diagnostic two: lead rich, demo poor

Symptom: 400 MQLs a month, 60 accepted by sales, 15 percent acceptance. Sales says marketing sends junk. Marketing says sales doesn’t work the leads.

Both are describing a definition failure. Score your last 200 MQLs on fit and behaviour separately and look at the distribution. If most crossed the threshold on behaviour alone, you’re passing people who read a lot of blog posts and work at companies that will never buy. Behaviour only scoring is the single most common cause of a low acceptance rate.

The fix is a hard fit gate before any behavioural score applies. Company size, region and industry as a pass or fail, then behaviour as the ranking within the qualified set. Expect MQL volume to drop 40 to 60 percent and acceptance to roughly double. The absolute number of opportunities usually goes up, but you have to warn the person who reports MQL volume to the board before you do it.

Diagnostic three: demo rich, opportunity poor

Symptom: plenty of meetings booked, few becoming opportunities. Reps are busy and the forecast is thin.

Usually this means qualification is happening in the meeting rather than before it. Every no budget, no authority, no timeline conversation costs a rep 45 minutes plus preparation. At 30 meetings a month with a 40 percent qualification rate, that’s roughly 18 wasted meetings, or two full selling days a month.

Two fixes, and I’d do both. Put qualifying questions on the booking form: company size, current tool, timeline. Then route accordingly, which is what tools like Chili Piper exist for. The second fix is call review. Listen to five disqualified calls in Gong and you’ll usually find the same unmet criterion repeating, which tells you exactly which targeting assumption is wrong.

The instrumentation this requires

You cannot run any of the above without four things in place.

A source field captured at first touch that persists to the opportunity record. Not last touch, not a re-stamped value. If a lead’s source changes when they later click an email, your channel analysis is fiction.

A timestamped event for every stage transition, stored on the record. Stage transition dates are what let you cohort properly and what let you measure velocity, which is the second most useful diagnostic after conversion rate.

Cohorting by entry week. Most CRM reports default to a close date view, which mixes entry cohorts and hides every trend.

And one named owner per stage definition, with the definitions written in a document both marketing and sales have read. That sounds like process theatre. It’s the thing that decides whether any of the numbers mean anything twelve months from now.

Segment and HubSpot or Salesforce will cover the mechanics. The harder part is the agreement, and no tool sells that. Wider context on the sources feeding this funnel sits in SaaS lead generation and in the ranked view at SaaS lead generation strategies, ranked.

What healthy looks like at 8 million ARR

A concrete shape, so you have something to compare against. 25,000 monthly sessions, 2.4 percent hand raiser rate giving 600 hand raisers, 35 percent to MQL giving 210, 55 percent accepted giving 115 SALs, 35 percent to SQL giving 40, and 60 percent to opportunity giving 24 opportunities a month.

At a 25 percent win rate and 25,000 dollars ACV, that’s six new customers and 150,000 dollars in new ARR a month. Change any single ratio by a fifth and watch what happens to the bottom line. That sensitivity is the argument for fixing stages rather than adding channels. Run your own version with the cost per lead calculator and compare against SaaS funnel conversion benchmarks and B2B SaaS funnel conversion benchmarks.

Do this first

Write the six definitions down this week and get a sales leader to sign off on them. Then run the cohort analysis on a month from at least 90 days ago and find your worst ratio.

You’ll probably discover the MQL to SAL number nobody has been reporting, and that it’s the one costing you the pipeline. What to do next depends on your deal size, which is why the tactics split cleanly by band in the lead generation playbooks by ACV band.